Proceedings Article10.1109/DCC52660.2022.00039
A Physics-Informed Vector Quantized Autoencoder for Data Compression of Turbulent Flow
M. Momenifar,Enmao Diao,Vahid Tarokh,Andrew D. Bragg +3 more
- 10 Jan 2022
pp 01-10
TL;DR: A physics-informed Deep Learning technique based on vector quantization is applied to generate a discrete, low-dimensional representation of data from simulations of three-dimensional turbulent flows to improve the CR, and reduce the mean square error of the MSE by an order of magnitude.
read more
Abstract: Analyzing large-scale data from simulations of turbulent flows is memory intensive, requiring significant resources. This major challenge highlights the need for data compression techniques. In this study, we apply a physics-informed Deep Learning technique based on vector quantization to generate a discrete, low-dimensional representation of data from simulations of three-dimensional turbulent flows. The deep learning framework is composed of convolutional layers and incorporates physical constraints on the flow, such as preserving incompressibility and global statistical characteristics of the velocity gradients. The accuracy of the model is assessed using statistical, comparison-based similarity and physics-based metrics. The training data set is produced from Direct Numerical Simulation of an incompressible, statistically stationary, isotropic turbulent flow. The performance of this lossy data compression scheme is evaluated not only with unseen data from the stationary, isotropic turbulent flow, but also with data from decaying isotropic turbulence, and a Taylor-Green vortex flow. Defining the compression ratio (CR) as the ratio of original data size to the compressed one, the results show that our model based on vector quantization can offer CR = 85 with a mean square error (MSE) of $O$(10−3), and predictions that faithfully reproduce the statistics of the flow, except at the very smallest scales where there is some loss. Compared to the recent study based on a conventional autoencoder where compression is performed in a continuous space, our model improves the CR by more than 30 percent, and reduces the MSE by an order of magnitude. Our compression model is an attractive solution for situations where fast, high quality and low-overhead encoding and decoding of large data are required.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Quadconv: Quadrature-Based Convolutions with Applications to Non-Uniform Pde Data Compression
TL;DR: QuadConv as mentioned in this paper is an approximation to continuous convolution via quadrature, which can be used on non-uniform, mesh-based data, and accomplishes this by learning a continuous kernel that can be sampled at arbitrary locations.
QCNN: Quadrature Convolutional Neural Network with Application to Unstructured Data Compression
TL;DR: Quadrature Convolutional Neural Networks (QCNNs) as discussed by the authors is an approximation to continuous convolution via quadrature, which can match the performance of standard discrete convolutions on structured uniform data, as in CNNs.
A Novel Side-Channel Archive Framework Using Deep Learning-based Leakage Compression
S.L. Jung,Sunghyun Jin,Hee Seok Kim +2 more
TL;DR: This paper proposes a novel side-channel data compression technique using autoencoders, offering higher compression rates than Deflate while maintaining decompression times and preserving data characteristics, verified through correlation power analysis experiments.
Statistical Deep Learning for Spatial and Spatio-Temporal Data
TL;DR: An overview of traditional statistical and machine learning perspectives for modeling spatial and spatio-temporal data is presented, and a variety of hybrid models that have recently been developed for latent process, data, and parameter specifica-tions are focused on.
References
•Proceedings Article
Neural Discrete Representation Learning
Aaron van den Oord,Oriol Vinyals,Koray Kavukcuoglu +2 more
- 02 Nov 2017
TL;DR: The Vector Quantised-Variational AutoEncoder (VQ-VAE) as discussed by the authors is a generative model that learns a discrete latent representation by using vector quantization.
•Posted Content
Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
TL;DR: This work considers a small-scale version of {\em conditional computation}, where sparse stochastic units form a distributed representation of gaters that can turn off in combinatorially many ways large chunks of the computation performed in the rest of the neural network.
3.3K
The gravitational settling of aerosol particles in homogeneous turbulence and random flow fields
TL;DR: In this article, the average settling velocity in homogeneous turbulence of a small rigid spherical particle subject to a Stokes drag force was shown to depend on the particle inertia and the free-fall terminal velocity in still fluid.
1.1K
•Posted Content
Generating Diverse High-Fidelity Images with VQ-VAE-2
TL;DR: It is demonstrated that a multi-scale hierarchical organization of VQ-VAE, augmented with powerful priors over the latent codes, is able to generate samples with quality that rivals that of state of the art Generative Adversarial Networks on multifaceted datasets such as ImageNet, while not suffering from GAN's known shortcomings such as mode collapse and lack of diversity.
1.1K
An inequality concerning the production of vorticity in isotropic turbulence
TL;DR: An inequality involving the rate of production S of mean-square vorticity in isotropic turbulence and a factor γ which may be said to allow for intermittency or for the non-vanishing of fourth-order cumulants was shown in this paper.
322